Maintaining electrical devices and equipment within specified temperatures is an important requirement for maintaining the operability of those devices. In the field of microelectronics, for example, microchips such as microprocessors must be maintained at or below maximum temperatures during operation to prevent the self-destruction of the microchips. Recent and continuing increases in the processing power and operating speeds of microchips has led to increases in the power dissipation of those microchips. For example, new high-powered chips can dissipate at least 10 Watts of power per chip and have heat fluxes as high as 500 W/cm2. Consequently, the cooling of microchips during their operation is becoming even more critical as the performance of those microchips becomes even greater.
Numerous techniques for cooling currently exist. With respect to the cooling of microchips in particular, many existing cooling techniques incorporate large and costly heat spreaders and heat sinks, which impose limits on package size and functionality. Other techniques involve directing one or more jets or streams of cool air or other gas (or liquid) at or along the one or more microchips that require cooling.
Referring to FIGS. 1 and 2, FIG. 1 shows a cross-sectional view of a prior art impinging jet system 20 for cooling microelectronic devices 22, and FIG. 2 shows a series of charts 24, 26, 28 representing airflow variation and vortex formation over time in the cross-sectional view of the prior art impinging jet system 20 cooling microelectronic devices 22. Impinging jet system 20 includes a jet 30 of fluid or coolant, such as air, that is generated by a pumping device 32, such as a fan or blower. Pumping device 32 drives jet 30 through an opening 34 in a material layer 36 spaced apart from an underlying target element, such as a circuit board 38. Jet 30 is directed toward circuit board 38, which supports a variety of heat generating elements such as microelectronic devices 22. In one example, jet 30 is generally directed toward one of microelectronic devices 22, referred to herein as a target 40, along an axis 42.
Jet 30 of impinging jet system 20 flows in an unsteady laminar flow mode. More particularly, jet 30 “buckles” or oscillates back and forth about axis 42 so that its central core is not always parallel to axis 42. This oscillation occurs in part due to the existence of vortices 44 (shown in FIG. 2). Vortices 44 are areas of high recirculation near opening 34 which can distort jet 30, entrain or remove some of the cool air from jet 30 exiting opening 34, and eliminate warmer air at the top of the space between material layer 36 and circuit board 38 where air exits the space. Due to this oscillation, jet 30 varies in its position relative to axis 42. In addition, a phenomenon related to the buckling occurs in which a central tip, or stagnation point 46, of jet 30 “sweeps” or moves back and forth along its target, e.g., target 40. This sweeping is represented in FIG. 1 by a bi-directional arrow 48. This sweeping is further represented in FIG. 1 by showing multiple axes 42 about which jet 30 oscillates, each of multiple axes 42 representing a different instant in time.
Stagnation point 46 is the point at which greatest pressure of jet 30 is provided against microelectronics devices 22, and is consequently where the greatest amount of cooling and maximum heat transfer coefficients occur. As a result of the sweeping motion of the tip of jet 30 across target 40, the position of stagnation point 46 moves back and forth relative to target 40 such that the portion of target 40 receiving the greatest amount of cooling varies in time. Such a phenomenon, referred to herein as an uncontrolled (or un-forced) jet technique, yields effective local cooling at target 40 but less uniform cooling for some remaining microelectronic devices 22 that also have high heat fluxes.
In order to provide cooling over greater areas, prior art solutions entail directing multiple steady (i.e., non-buckling) impinging jets 30 parallel to one another from respective rectangular openings 34 in material layer 36. By using multiple steady jets 30, the heat transfer coefficient may be maintained at a more uniform level over a wider region of circuit board 38, over two (or more) microelectronic devices 22. As such the temperature along a top surface of microelectronic devices 22 and circuit board 38 may be reduced over a larger area.
However, a multiple jet impinging system may still produce non-uniformities in heat transfer across the surface of circuit board 38 thus leading to inefficiencies in cooling. Typically, material layer 36 is a circuit board with its own microelectronic devices coupled to it. Thus, multiple openings 34 in such a circuit board utilize an undesirably large amount of space that could be better served for the attachment of devices and the formation of electrically conductive traces. In addition, the inclusion of multiple openings 34 calls for additional design and manufacturing processes which drives up cost. Furthermore, the multiple openings 34 may necessitate additional pumping devices 32 or structure to direct coolant flow to the multiple openings 34, further driving up costs.
Thus, what is needed is a technique for effectively cooling microelectronic devices which provides a generally uniform cooling effect. What is further needed is a technique that can be cost effectively implemented and can provide flexibility in cooling capabilities.